Abstract This proposal seeks to establish characteristics of the mathematical models describing the progressions of salivary progesterone and estradiol hormone levels through the course of a woman's menstrual cycle. Specifically, time and magnitude of peak hormone velocities will be estimated using derivatives of the mathematical models developed. Large sample variances of these estimates will be obtained. Using these characteristics, Monte-Carlo simulations will be performed to determine the capacity of our models in estimating time and magnitude of the peak hormone rates. Simulations will include several realistic designs such as data from complete cycle (28 days), missing 10% at random and a truncated set of 14 contiguous days of measurements. The methods developed will be applied to data collected from a previous study with a primary focus of testing hypotheses regarding the impact of the characteristics of the models mentioned above, on reactivity to stressful and smoking related cues and daily smoking behavior.